1. Technical Field
The present invention relates to a musical sound producing apparatus, a musical sound producing method, a musical sound producing program, and recording medium for automatically producing musical sound data corresponding to image data.
2. Background Art
As a technique which controls playing corresponding to an image, for example, Japanese Patent 2629740 discloses a technique which controls tempo or the like by making use of a profile of an object to be photographed. In this technique, respective signals of R (red), G (green), B (blue) are separated from inputted video signals, and gray scale data indicative of gray scales are generated as digital data for respective colors. Then, the object to be photographed is specified based on the gray scale data of respective colors and preset threshold value data thus detecting the profile of the object to be photographed, and the playing is controlled corresponding to “the complexity of the detected profile”.
Japanese Laid-open Patent Publication 2002-276138 discloses a technique which produces musical sound by detecting a position of a moving manipulation object, wherein the position of the specified manipulation object having a fixed shape is detected, and musical sounds are generated corresponding to both elements consisting of a traveling time from an arbitrary position to a current position of the manipulation object and the current position. To be more specific, when a position of a specified portion of the object to be photographed is detected, musical sound which is produced is allocated to a sound producing region set on an image display screen, and after a lapse of a predetermined time from the determination that the specified portion is not present in one region on the image display screen, it is determined that the specified portion exists in another region on the different image display screen, and the determined another region belongs to the sound producing region, the musical sound allocated to the sound producing region is generated.
On the other hand, as a technique which overcomes a problem which arises in the production of musical sound by catching the movement of an object, for example, Japanese Laid-open Patent Publication 2000-276139 discloses a technique in which a plurality of motion vectors is extracted from each block of a supplied image, one control vector is calculated from the plurality of motion vectors, and musical sound is produced based on the calculated control vector.
In the method which extracts the plurality of motion vectors from each block of the image, in respective blocks (16×16) corresponding to a specified image frame and an image frame which follows the specified image frame, pixels which exhibit the least color difference are picked up and the difference of positions of these pixels is set as the motion vector.
However, in the technique disclosed in Japanese Patent 2629740, it is necessary to determine the complexity of a profile of an object to be photographed by using a still image as an object, by decomposing color signals of the still image, specifying the object to be photographed by threshold inspections for respective colors, and by detecting a profile of the object to be photographed. Accordingly, this technique is an existing sound data modifying technique in view of a drawback that a load of processing is increased and the complexity of the profile. Accordingly, Japanese Patent 2629740 has a drawback that the patent has no idea of producing musical sound.
The technique disclosed in Japanese Laid-open Patent Publication 2000-276138 discloses the judgment on the movement which follows a registered specified operator and aims at the production of musical sound. However, this technique has a drawback that musical sound is not produced from an arbitrary motion picture frame.
The technique disclosed in Japanese Laid-open Patent Publication 2000-276139 copes with a task to produce musical sounds based on the analysis of the motion and also develops a method which detects motion vectors by performing the analysis in a limited specified region for reducing a load on the analysis. However, this technique is a technique which cannot avoid a fundamental drawback that a large load is applied to the calculation of the motion vectors.
It is an object of the present invention to provide a technique which, using continuous motion picture frames as objects, can take out motion data using a simple method and can produce musical sound data based on this taken-out motion data. It is also an object of the present invention to construct a unique application field by further combining the musical sound data produced in such a manner with an existing technique.
Accordingly, the present invention provides a musical sound producing apparatus, a musical sound producing method, a musical sound producing program and a recording medium for automatically producing musical sound data by calculating motion data based on inputted image data using a simple technique without preparing playing information or the like in advance.